提交 fed66381 编写于 作者: E Eric Dumazet 提交者: David S. Miller

net: pskb_expand_head() optimization

Move frags[] at the end of struct skb_shared_info, and make
pskb_expand_head() copy only the used part of it instead of whole array.

This should avoid kmemcheck warnings and speedup pskb_expand_head() as
well, avoiding a lot of cache misses.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
上级 c1f79426
......@@ -202,10 +202,11 @@ struct skb_shared_info {
*/
atomic_t dataref;
skb_frag_t frags[MAX_SKB_FRAGS];
/* Intermediate layers must ensure that destructor_arg
* remains valid until skb destructor */
void * destructor_arg;
/* must be last field, see pskb_expand_head() */
skb_frag_t frags[MAX_SKB_FRAGS];
};
/* We divide dataref into two halves. The higher 16 bits hold references
......
......@@ -817,7 +817,7 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
memcpy(data + nhead, skb->head, skb->tail - skb->head);
#endif
memcpy(data + size, skb_end_pointer(skb),
sizeof(struct skb_shared_info));
offsetof(struct skb_shared_info, frags[skb_shinfo(skb)->nr_frags]));
for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
get_page(skb_shinfo(skb)->frags[i].page);
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册